guardrails-ai (PyPI)

Known vulnerabilities and malicious package reports for guardrails-ai on PyPI, live from OSV, with CISA KEV exploitation status and EPSS scores.

Known package reports

1 malicious package report and 4 vulnerabilities returned by OSV. Check a specific version to see which reports apply.

If you installed a reported malicious release, remove it and rotate the credentials that machine could access. Upgrading does not undo code that already ran.

MAL-2026-3607: Malicious package report

Malicious code in guardrails-ai (PyPI)

Published .

Updated .

Reported affected releases: 0.10.1.

GHSA-r6hf-g5x6-7pv9: Vulnerability

Guardrails AI contains a code injection vulnerability in its Hub package installation mechanism

Published .

Updated .

Affected ranges: introduced 0, last_affected 0.6.7.

Severity: CRITICAL.

CVE-2026-31233

GHSA-xmpw-2vmm-p4p6: Vulnerability

Malicious code in guardrails-ai 0.10.1 (supply chain compromise)

Published .

Updated .

Reported affected releases: 0.10.1.

Severity: CRITICAL.

CVE-2026-45758

GHSA-f8hx-f4xw-c646: Vulnerability

Guardrails AI vulnerable to Improper Restriction of XML External Entity Reference

Published .

Updated .

Affected ranges: introduced 0, fixed 0.5.0.

Fix versions reported by the advisory: 0.5.0.

Severity: MEDIUM.

CVE-2024-6961

GHSA-w392-75q8-vr67: Vulnerability

Guardrails has an arbitrary code execution vulnerability

Published .

Updated .

Affected ranges: introduced 0.2.9, fixed 0.5.10.

Fix versions reported by the advisory: 0.5.10.

Severity: HIGH.

CVE-2024-45858

Source: OSV, including OpenSSF malicious-package reports and GitHub advisories. No known report is not proof a package is safe.